Counterpoint: There is not capillary recruitment in active skeletal muscle during exercise.

The notion that a substantial proportion of capillaries do not contain moving red blood cells (RBCs) in muscle at rest but are “recruited,” i.e., begin flowing with RBCs during contractions, is one basis for our present understanding of blood-muscle exchange during exercise (20, 28). This concept emanates, in part, from August Krogh, who showed that many capillaries in resting muscle did not contain India ink after high pressure perfusion (19). Despite Krogh himself recognizing that India ink particles clumped together, more likely to prevent complete perfusion of the capillary bed at rest than during exercise, these experiments, and Krogh’s O2 diffusion model based on them, are still cited by researchers invoking capillary recruitment (e.g., Refs. 22). Today capillary recruitment during exercise is accepted by many to explain important physiological phenomena, including: 1) greater blood-muscle delivery and extraction of O2, free fatty acids, and glucose and 2) reduced capillaryto-mitochondrial diffusion distances. It makes great sense that, if there were a reserve of capillaries at rest, during exercise when the muscle demands for O2 may increase up to 100-fold, all—or at least most—capillaries would contribute to meet that demand. Why, therefore, choose to oppose the concept of capillary recruitment during exercise? In Britain, the motto of The Royal Society is “Nullius in Verba” (Take nobody’s word for it, see it for yourself). However, the majority of research papers invoking capillary recruitment have not visualized the capillary bed (e.g., Refs. 3, 7, 25). In his letter to the editor of The American Journal of Physiology, the eminent microcirculation expert, Professor Eugene Renkin (26) criticized Dr. Bentzer for entitling his paper “Capillary filtration coefficient is independent of number of perfused capillaries in skeletal muscle”(4) on the basis that “. . . its title is misleading. No direct measurements of [RBC-flowing] capillary number were made. . . .” The same criticism could be leveled at almost all papers that include the words capillary recruitment in their titles. Evidence that most capillaries already sustain flow in resting muscle would preclude the possibility that recruitment of previously nonperfused capillaries occurs to any great extent during contractions. Direct evidence for RBC flow in most capillaries in resting muscle. In resting muscle, intravital light microscopy shows that over 80% of capillaries support RBC flow, e.g., in rat spinotrapezius (14, 17, 24), diaphragm (15), and extensor digitorum longus (1), hamster cremaster and sartorius (8), cat sartorius (6), rabbit tenuissimus (30). However, animals in these experiments were anesthetized to facilitate muscle exteriorization and viewing of the capillary beds. To address this, Bailey and colleagues (2) employed minimally invasive techniques to measure blood flow (radioactive microspheres) and microvascular oxygen partial pressure in muscle in situ and neither criterion was altered by exteriorization. Moreover, the dynamic matching of increased O2 delivery and V̇O2 during contractions in situ was preserved in the exteriorized muscle. It is difficult to conceive how anesthesia might affect arteriolar smooth muscle function at rest (to produce a falsely high %RBC-perfused capillaries) and yet muscles increase their blood flow and V̇O2 at a ratio of 6:1 (11), which is precisely that seen in intact voluntary exercising animals and humans (23). Another valid concern about intravital microscopy is that there are non-RBC containing capillaries at rest that cannot be seen because of their translucency. However, neither observation of contracting (17) nor vasodilated (16) muscles revealed a significant number of such vessels (see also Refs. 8, 14). The technical requirements necessary to observe capillaries within living muscle restrict the procedure to a limited selection of animal muscles so that one question to ask is: How representative of other muscles in the animals’ body and in humans are these? In anesthetized and conscious animals, Snyder et al. (29) used systemic indicator injections and demonstrated that essentially all capillaries in each muscle examined (vastus lateralis, diaphragm, soleus) were perfused within 3–7 s. Indirect evidence for RBC flow in most capillaries in resting human muscle. Noninvasive near-infrared spectroscopy (NIRS) measures muscle hemoglobin concentration ([Hb]). If there were significant recruitment of previously non-RBC containing capillaries during exercise, say from 20 to 90%, [Hb] would be expected to increase several-fold. However, the rest-to-exercise [Hb] increase is less than onefold (e.g., 10) and can be accounted for by increased capillary hematocrit (18). Thus, as in animal muscles, there is little room for substantial capillary recruitment in human muscle. Against the evidence for capillary recruitment during exercise. The literature that purports to demonstrate capillary recruitment deserves to be evaluated on its own merits, but the following must be considered as possible explanations for reports of many non-RBC flowing capillaries in resting muscle. 1) Capillaries are fragile structures, subject to damage by blunt trauma, surgery, and/or manipulations such as stretching (24). 2) PO2 within resting muscle is normally very low and raising this will cause arteriolar constriction and capillary flow stoppage (22). 3) Anesthetized preparations are often hypovolemic and hypotensive, which provokes reflex vasoconstriction. In addition, misinterpretation of histological techniques has supported the notion of capillary recruitment. The conclusion that a RBC in the muscle capillary cross-section indicates RBC flow, whereas its absence supports no flow is erroneous (e.g., Ref. 13). In resting animal muscle observed in vivo, RBC movement in flowing capillaries varies over time, appearing either continuous or stop-start (9). Muscle contractions result Point:Counterpoint